Toggle light / dark theme

From insilico meddicine — the beginning of an AI healthcare revolution.


Poly Mamoshina on Machine Learning for small molecule drug discovery and the beginning of an AI healthcare revolution — interviewed at the Undoing Aging conference in Berlin 2019!

Polina Mamoshina is a senior research scientist at Insilico Medicine, Inc (www.insilico.com), a Baltimore-based bioinformatics and deep learning company focused on reinventing drug discovery and biomarker development and a part of the computational biology team of Oxford University Computer Science Department. Polina graduated from the Department of Genetics of the Moscow State University. She was one of the winners of GeneHack a Russian nationwide 48-hour hackathon on bioinformatics at the Moscow Institute of Physics and Technology attended by hundreds of young bioinformaticians. Polina is involved in multiple deep learning projects at the Pharmaceutical Artificial Intelligence division of Insilico Medicine working on the drug discovery engine and developing biochemistry, transcriptome, and cell-free nucleic acid-based biomarkers of aging and disease. She recently co-authored seven academic papers in peer-reviewed journals.

A new study shows that short-term treatment with the common organ rejection drug rapamycin reverses periodontal bone loss, attenuates inflammation, and makes the oral microbiome revert to a more youthful state in old mice.

What is rapamycin?

Rapamycin (also known as sirolimus) is a macrolide, a class of antibiotics that includes Biaxin (Clarithromycin), Zithromax (Azithromycin), Dificid (Fidoximycin), and Erythromycin. Macrolides inhibit the growth of bacteria and are often used in the treatment of common bacterial infections.

Nearly one in six deaths from prostate cancer could be prevented if targeted screening was introduced for men at a higher genetic risk of the disease, according to a new UCL-led computer modelling study.

Prostate cancer is the most common form of cancer in men with around 130 new cases diagnosed in the UK every day and more than 10,000 men a year dying as a result of the disease. However, unlike breast and there is currently no national programme for this disease in the UK.

A blood test that detects raised levels of the prostate-specific antigen (PSA) can be used to screen for . However, this test is not a reliable indicator as it does not accurately distinguish between dangerous cancers from harmless ones—leading to both unnecessary operations and missed cancers that are harmful.

Every year, we’re reminded to return to the pharmacy for a flu shot. Why can’t we have a flu vaccine that offers long-term protection, like those for measles or polio? That’s because the influenza virus continuously evolves, so the immune response we build up one year might not work the next year—or even on the version of the flu you catch the same year. As a result, the virus remains dangerous: last year, it caused more than 60,000 deaths in the United States alone.

New findings, published in Cell, reveal why making a general-purpose vaccine that guards against all versions of the flu is so hard: Instead of improving its memory of the previous version of virus, the develops its response to the new virus variant from scratch, mostly using that have no memory of the virus.

“If we can figure out how to help the immune system to keep building on what it has already learned, we could develop better vaccines for highly evolving viruses like the flu, or HIV, or Hepatitis C,” says Gabriel D. Victora, assistant professor at Rockefeller.

P53 is the most famous cancer gene, not least because it’s involved in causing over 50% of all cancers. When a cell loses its p53 gene—when the gene becomes mutated—it unleashes many processes that lead to the uncontrolled cell growth and refusal to die, which are hallmarks of cancer growth. But there are some cancers, like kidney cancer, that that had few p53 mutations. In order to understand whether the inactivation of the p53 pathway might contribute to kidney cancer development, Haifang Yang, Ph.D., a researcher with the Sidney Kimmel Cancer Center—Jefferson Health probed kidney cancer’s genes for interactions with p53.

Earlier work found that PBRM1—the second most mutated gene in —could interact with p53. However, other researchers were unable to definitively show that it was truly an important mechanism in kidney cancer.

Rather than looking at the p53 protein itself, first author Weijia Cai a postdoc in Dr. Yang’s lab and other collaborators looked at an activated version of p53, one that is studded with an additional chemical marker—an —at many specific spots.

Anxious Astronaut has suffered an anxiety attack in space. It could be debilitating, they’re not sure. And unlike, say, a broken arm, it is not immediately visible to Anxious’ co-workers. Anxious Astronaut is good at hiding their problem, which is how they got through the screening process on Earth. But Anxious Astronaut needs to be operating at peak functionality, which Anxious Astronaut knows, which is making them more stressed, and they haven’t even acknowledged to themselves that they’re undergoing a silent crisis. Stress is tough.

Anxious Astronaut does not want to give up their duties, so they’re not taking time to self-evaluate. And besides, any human diagnosis is millions of miles of way, considering Anxious Astronaut and their team are halfway to Mars. So how can Anxious Astronaut’s team figure out what’s wrong? A biosensor. A small, nearly invisible biosensor placed on Anxious Astronaut’s forehead has detected unusually high cortisol, which the body releases when stressed. The data is shared with the medical staff on the mission, and Anxious is able to have their workloads reduced until they’re feeling up to snuff.

Thanks to developments in biosensors that NASA and outside group NextFlex are working on today, Anxious or Unhealthy Astronaut might be able to figure out what’s ailing them at speeds unimaginable today.

Any comments?


Ultraprecise 3D printing technology is a key enabler for manufacturing precision biomedical and photonic devices. However, the existing printing technology is limited by its low efficiency and high cost. Professor Shih-Chi Chen and his team from the Department of Mechanical and Automation Engineering, The Chinese University of Hong Kong (CUHK), collaborated with the Lawrence Livermore National Laboratory to develop the Femtosecond Projection Two-photon Lithography (FP-TPL) printing technology.

By controlling the spectrum via temporal focusing, the laser 3D printing process is performed in a parallel layer-by-layer fashion instead of point-by-point writing. This new technique substantially increases the printing speed by 1,000—10,000 times, and reduces the cost by 98 percent. The achievement has recently been published in Science, affirming its technological breakthrough that leads nanoscale 3D printing into a new era.

The conventional nanoscale 3D , i.e., two-photon polymerization (TPP), operates in a point-by-point scanning fashion. As such, even a centimeter-sized object can take several days to weeks to fabricate (build rate ~ 0.1 mm3/hour). The process is time-consuming and expensive, which prevents practical and industrial applications. To increase speed, the resolution of the finished product is often sacrificed. Professor Chen and his team have overcome the challenging problem by exploiting the concept of temporal focusing, where a programmable femtosecond light sheet is formed at the focal plane for parallel nanowriting; this is equivalent to simultaneously projecting millions of laser foci at the , replacing the traditional method of focusing and scanning laser at one point only. In other words, the FP-TPL technology can fabricate a whole plane within the time that the point-scanning system fabricates a point.

From the beginning of time, humankind has searched for the secret to a long life. Now science may have found an answer, in the form of molecular augury. The pattern of chemical chains that attach to the DNA in your cells—on-off switches known as epigenetic markers—can reveal how swiftly you are aging, and perhaps even how much longer you will live. While genetic testing might tell you where you came from, epigenetics promises a glimpse into the future. Now, a handful of companies are offering commercial blood or saliva tests based on the science of epigenetics—a chance to find out how old you truly are.


Companies claim they can now easily calculate your biological age. Should you take them up on it?